Perceptual load theory (Lavie & Tsal, 1994) proposes that our attentional selection is more efficient under conditions of high perceptual load due to spare capacity being insufficient to “spillover” and fully process nontarget flankers. Tsal and Benoni (2010) propose that reductions in flanker interference are due to dilution rather than a lack “spilled over” attentional resources. Gaspelin, Ruthruff, and Jung (2014) have suggested that neither perceptual load theory nor dilution adequately explain the effects of nontarget stimuli on flanker interference. Their “slippage” account proposes that a visually-similar, to-be-ignored flanker involuntarily captures attention on a proportion of the trials.
Ninety participants were randomly assigned to 6 conditions resulting from a 3 (load: low, high, low with dilution) X 2 (cue-target SOA: 0 or 100 ms) X 3 (target-flanker distance: 1-3 positions) X 2 (target-flanker compatibility: compatible vs incompatible) mixed design, with load and SOA being between-subjects factors.
Participants viewed a fixation display consisting of a center dot surrounded by a circular array of 7 dots for 1000 ms. In the low load condition, two of the dots on the circle were replaced by the letter T rotated 90 deg clockwise or anticlockwise. The target was marked by a cue consisting of four dots. The unmarked flanking T could be placed 1-3 positions away on the circle, and could be rotated in the same direction of the target (compatible) or in the opposite direction (incompatible). In the high load and low mirrored L shapes, which were a contrasting green in the dilution condition.
In all conditions, flanker interference diminished with cue-target separation. The overall magnitude of flanker interference was highest in the low load condition, but was significantly modulated by a 114 ms SOA precue at the two-position separation. The overall magnitude of flanker interference was lower in the high load condition, but the pattern of cue modulation was similar. In the low load with dilution condition. Our results are consistent with those of Gaspelin et al. (2014) in that neither perceptual load nor dilution fully account for the results. Flanker interference is also modulated by the precue as predicted by slippage theory.